Abstract

Accelerograms recorded near active faults have some important characteristics that make them different from those recorded in far-fault regions. High-frequency components in acceleration records and long-period velocity pulses are among notable specifications of such ground motions. In this paper, a moving average filtering with appropriate cut-off frequency has been used to decompose the near-fault ground motions into two components having different frequency contents: first, Pulse-Type Record (PTR) that possesses long-period pulses; second, the relatively high-frequency BackGround Record (BGR), which does not include large velocity pulses. Comparing the results with those extracted through wavelet analysis shows that moving average filter is an appropriate and efficient tool for near-fault records decomposition. The method is applied to decompose a suite of 91 selected near-fault records and the elastic response of structures is examined through their response to the decomposed parts. The results emphasizes that in contrast with ordinary far-fault earthquake records, response spectra of near-fault ground motions typically have two distinct local peaks, which are representatives of the high- and low-frequency components, i.e., BGR and PTR, respectively. Moreover, a threshold period is identified below which the response of structures is dominated by BGR while PTR controls the response of structures with periods longer than this period.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call